"Forward-only" algorithms, which train neural networks while avoiding a
...
The ability of deep neural networks to generalise well even when they
in...
Learning rate schedules are ubiquitously used to speed up and improve
op...
From the sampling of data to the initialisation of parameters, randomnes...
Autoencoders are the simplest neural network for unsupervised learning, ...
A recent series of theoretical works showed that the dynamics of neural
...
Direct Feedback Alignment (DFA) is emerging as an efficient and biologic...
Contact-tracing is an essential tool in order to mitigate the impact of
...
Deep neural networks can achieve remarkable generalization performances ...